The study of neuronal activity in awake, freely moving animals is the most representative view of 'normal' neuronal function. Anesthetics and restraint have profound effects on neuronal physiology and the correlate animal behavior. The study of real time neuronal event processing requires recording methods with high temporal bandwidth (>500 hz) and the ability to detect physiologically relevant events (i.e. voltage changes). Currently only microelectrode recording of single neuronal units, multiunit activity and field potentials have sufficient temporal resolution and portability to allow recording of neuronal activity in freely moving animals. The proposed project will produce three miniature microscope/imaging systems which can be head-mounted and will allow the optical recording of changes in membrane voltage of neurons in freely moving rats. During the funded two-year ARRA project, two prototype devices and two image sensors were developed that collect wide field image sequences of fluorescent voltage dye signals. The current application will complete the fabrication and operationalize two different style, head-mountable, high-speed, fluorescence microscopes. This includes the design and fabrication of two different finalized, imaging sensors, custom optics for each, a stabilized illumination source and form fit microscope bodies with skull mounts. The studies proposed herein will also build and test a novel fluorescence microscope design that will reduce the height of the traditional microscope by eliminating the 45 dichroic mirror. We will also miniaturize our current camera control and recording system to be fully contained in a back pack on the rat. This back pack will contain a field programmable gate array, heat exchanger/coolant pump, SD memory and a battery to run the microscope fully autonomously. Finally, the instruments will be tested in imaging studies of the rat somatosensory cortex (barrel cortex). Cortical responses to object discrimination will be studied in freely moving rats. These studies will systematically engineer each component to maximize excitation light delivery, fluorescent light collection and recording speed while minimizing weight volume, energy usage and heat generation. The device will be capable of recording rapid (1000 fps) fluorescent image sequences of a >2 mm2 area of cortex and detect small changes in ?F/F (0.1%). The device will be small (<1 cm2 head mounted) and light weight enough (5 g) to be mounted on the head of a freely moving rat. The device is intended as a precursor to devices that can translate neuronal activity in humans into actions in real time, optical brain machine prosthetic.